A Study of the Relationships of Consumer Investment Sentiment and Stock Price:VAR Model

Autor: Yuan-ling Chen, 陳苑菱
Rok vydání: 2009
Druh dokumentu: 學位論文 ; thesis
Popis: 97
This article intends to study whether the correlation exists between Investors’ Sentiment and Taiwan Stock Market. The Investors’ Sentiment is substituted by Taiwan Consumer Confidence Index (CCI) compiled and issued by the Research Center for Taiwan Economic Development (RCTED) in National Central University; the reviews are conducted based on monthly data from 2001 to 2008, combined with indirect Investor Sentiment Index, Taiwan Stock Index and Industrial Production Index such as Native Investment Proportions and Stock Turnover. At the first stage, for the six subjective variables of six sub-index from Taiwan Consumer Confidence Index (CCI) compiled by the Research Center for Taiwan Economic Development (RCTED) in National Central University, the variable dimension is reduced through Factor Analysis, to extract the more representative variables in every factor. At the second stage, Vector Autoregressive Model (VAR) is applied to conduct short-term interactive tests. The results show that: at the first stage, Factor Analysis is applied in this article to reduce the variable dimension, which is effective to reduce the less quantity of dimension in the acquired items after being investigated, while able to keep most of the data provided by the original data. Through “Varimax” by Factor Analysis, the new variables produced after being varied are tested to see their correlation with Taiwan Stock Index, which is more significant that that of primitive sequence(the timing for stock investment in half a year later). Previously, most domestic studies adopted indirect Sentiment Index to measure Investor Sentiment Index, and adopted less the statistic results acquired from the investigation on investors’ ways. In this article, Taiwan “Consumer Confidence Index (CCI)” is chosen, able to be another Sentiment Index provided for studies of investor sentiment in the future.
Databáze: Networked Digital Library of Theses & Dissertations